1. Field of the Invention
The present invention relates to a region recognizing device for recognizing a region marked with a closed loop on an original document to be read.
2. Description of the Prior Art
Sometimes, it is necessary to eliminate a part of an original document, hereafter referred to as an original, or to shift a part of it to a different position. Heretofore, the original has been copied, the copy cut into pieces, and the pieces rearranged to form a desired original. However, this method is troublesome because it is necessary to cut the copy into pieces and to glue the pieces. Furthermore, in this method, steps are formed on the original along the border lines of the pieces glued. When the original is read, shadows sometimes form by these steps, causing noises representing the border lines to appear when the original is read mechanically.
A method of using a coordinates specifying device to specify a region of an original which is to be removed, shifted or so forth has been employed in the art. However, in this method, it is necessary to specify X-Y coordinates, and the X-Y coordinates specifying operation takes a relatively long period of time. Furthermore, a large number of points must be specified for intricate configurations. Thus, the method is applicable only to the specifying of simple configurations.
For the purpose of eliminating the above-described difficulties, an image processing system has been proposed by Japanese Patent Application (OPI) No. 194670/1982 (the term "OPI" means "an unexamined published application"). In such image processing system, an original in which a region is specified with a loop is optically read, the resultant optical image is utilized to obtain loop data, and the optically read data of the original are processed according to the loop data.
More specifically, the loop is drawn in a predetermined color and the optical reading operation is carried out with color discriminating means. The predetermined color data obtained is subjected to main scanning compression and auxiliary scanning compression. The color data thus compressed is supplied to an edge detector to detect edges. The detection data is supplied to a region detector comprising a D flip-flop. The flip-flop is set by the front edge data in a line, and reset by the rear edge data. Thus, the region specified can be recognized with a pair of front and rear edge data.
If, in the above-described conventional image processing system, more than one closed-loop is drawn on one original, then edge data are provided in mixed form. This sometimes results in a situation where it is impossible to accurately distinguish the closed loops from one another and, therefore, sometimes the region determining operation involves errors. In the case where an intricately configured closed loop is drawn, odd numbered pieces of edge data are sometimes read from a single line. In this case, the conventional system cannot accurately recognize the region, because it utilizes the edge data of the preceding line.
Accordingly, an object of this invention is to provide a region recognizing device which can recognize the region or regions with high accuracy even if a plurality of closed loops are drawn on an original or a complicated region is specified.